Systems, methods, and devices for determining an introduction portion in a video program

Abstract

Systems, methods, and devices relating to determining an introduction portion in a video program are described herein. A method may determine first and second hard-matching pairs of video segments in first and second video content such that video fingerprints of the first hard-matching pair match and video fingerprints of the second hard-matching pair also match. The method may classify a third pair of video segments in the first and second video content, sequentially between the first and second hard-matching pairs, as a soft-matching pair of video segments of an introduction portion. The method may use the classification of the third pair of video segments as a soft-matching pair to determine a model configured to determine that a pair of video segments in two video content items are a soft-matching pair of video segments of an introduction portion.

What is claimed is: 
     
         1 . A method comprising:
 classifying a pair of video segments in first video content and second video content as a soft-matching pair of video segments of an introduction portion of at least one of the first video content or the second video content, wherein the pair of video segments is sequentially between first and second hard-matching pairs of video segments; and   determining, based on the classifying the pair of video segments as a soft-matching pair of video segments of an introduction portion of at least one of the first video content or the second video content, a model configured to determine that a pair of video segments in two video content items are a soft-matching pair of video segments of an introduction portion of at least one of the two video content items.   
     
     
         2 . The method of  claim 1 , wherein the first video content comprises at least a portion of a first episode of a video program series and the second video content comprises at least a portion of a second episode of the video program series. 
     
     
         3 . The method of  claim 1 , wherein the first video content comprises target video content in which the introduction portion is not known and the second video content comprises reference video content in which the introduction portion is known. 
     
     
         4 . The method of  claim 1 , further comprising:
 determining the model via machine learning, wherein a training data input for the machine learning comprises video fingerprints of the classified pair of video segments in the first video content and the second video content, and a training data output for the machine learning comprises the classification of the pair of video segments as a soft-matching pair of video segments of the introduction portion of the at least one of the first video content or the second video content.   
     
     
         5 . The method of  claim 4 , wherein the model comprises a regressor model and a training data input for determining the regressor model comprises a difference between video fingerprints of the classified pair of video segments in the first video content and the second video content. 
     
     
         6 . The method of  claim 1 , wherein:
 a difference between lengths of the first hard-matching pair of video segments satisfies a length threshold, and   a difference between lengths of the second hard-matching pair of video segments satisfies the length threshold.   
     
     
         7 . The method of  claim 6 , wherein a difference between lengths of the classified pair of video segments in the first video content and the second video content does not satisfy the length threshold. 
     
     
         8 . The method of  claim 1 , wherein video fingerprints of the first hard-matching pair of video segments match, video fingerprints of the second hard-matching pair of video segments match, and video fingerprints of the classified pair of video segments in the first video content and the second video content do not match. 
     
     
         9 . The method of  claim 8 , wherein a video fingerprint of a video segment comprises an alphanumeric value, and a matching pair of video fingerprints each comprise the same alphanumeric value. 
     
     
         10 . A non-transitory computer-readable medium storing instructions that, when executed, cause:
 classifying a pair of video segments in first video content and second video content as a soft-matching pair of video segments of an introduction portion of at least one of the first video content or the second video content, wherein the pair of video segments is sequentially between first and second hard-matching pairs of video segments; and   determining, based on the classifying the pair of video segments as a soft-matching pair of video segments of an introduction portion of at least one of the first video content or the second video content, a model configured to determine that a pair of video segments in two video content items are a soft-matching pair of video segments of an introduction portion of at least one of the two video content items.   
     
     
         11 . The non-transitory computer readable medium of  claim 10 , wherein the first video content comprises at least a portion of a first episode of a video program series and the second video content comprises at least a portion of a second episode of the video program series. 
     
     
         12 . The non-transitory computer readable medium of  claim 10 , wherein the first video content comprises target video content in which the introduction portion is not known and the second video content comprises reference video content in which the introduction portion is known. 
     
     
         13 . The non-transitory computer readable medium of  claim 10 , wherein the instructions, when executed, further cause:
 determining the model via machine learning, wherein a training data input for the machine learning comprises video fingerprints of the classified pair of video segments in the first video content and the second video content, and a training data output for the machine learning comprises the classification of the pair of video segments as a soft-matching pair of video segments of the introduction portion of the at least one of the first video content or the second video content.   
     
     
         14 . The non-transitory computer readable medium of  claim 13 , wherein the model comprises a regressor model and a training data input for determining the regressor model comprises a difference between video fingerprints of the classified pair of video segments in the first video content and the second video content. 
     
     
         15 . The non-transitory computer readable medium of  claim 10 , wherein:
 a difference between lengths of the first hard-matching pair of video segments satisfies a length threshold, and   a difference between lengths of the second hard-matching pair of video segments satisfies the length threshold.   
     
     
         16 . The non-transitory computer readable medium of  claim 15 , wherein a difference between lengths of the classified pair of video segments in the first video content and the second video content does not satisfy the length threshold. 
     
     
         17 . The non-transitory computer readable medium of  claim 10 , wherein video fingerprints of the first hard-matching pair of video segments match, video fingerprints of the second hard-matching pair of video segments match, and video fingerprints of the classified pair of video segments in the first video content and the second video content do not match. 
     
     
         18 . The non-transitory computer readable medium of  claim 17 , wherein a video fingerprint of a video segment comprises an alphanumeric value, and a matching pair of video fingerprints each comprise the same alphanumeric value. 
     
     
         19 . A device comprising:
 one or more processors; and   memory storing instructions that, when executed by the one or more processors, cause the device to:   classify a pair of video segments in first video content and second video content as a soft-matching pair of video segments of an introduction portion of at least one of the first video content or the second video content, wherein the pair of video segments is sequentially between first and second hard-matching pairs of video segments; and   determine, based on the classifying the pair of video segments as a soft-matching pair of video segments of an introduction portion of at least one of the first video content or the second video content, a model configured to determine that a pair of video segments in two video content items are a soft-matching pair of video segments of an introduction portion of at least one of the two video content items.   
     
     
         20 . The device of  claim 19 , wherein the first video content comprises at least a portion of a first episode of a video program series and the second video content comprises at least a portion of a second episode of the video program series. 
     
     
         21 . The device of  claim 19 , wherein the first video content comprises target video content in which the introduction portion is not known and the second video content comprises reference video content in which the introduction portion is known. 
     
     
         22 . The device of  claim 19 , wherein the instructions, when executed by the one or more processors, further cause the device to:
 determining the model via machine learning, wherein a training data input for the machine learning comprises video fingerprints of the classified pair of video segments in the first video content and the second video content, and a training data output for the machine learning comprises the classification of the pair of video segments as a soft-matching pair of video segments of the introduction portion of the at least one of the first video content or the second video content.   
     
     
         23 . The device of  claim 22 , wherein the model comprises a regressor model and a training data input for determining the regressor model comprises a difference between video fingerprints of the classified pair of video segments in the first video content and the second video content. 
     
     
         24 . The device of  claim 19 , wherein:
 a difference between lengths of the first hard-matching pair of video segments satisfies a length threshold, and   a difference between lengths of the second hard-matching pair of video segments satisfies the length threshold.   
     
     
         25 . The device of  claim 24 , wherein a difference between lengths of the classified pair of video segments in the first video content and the second video content does not satisfy the length threshold. 
     
     
         26 . The device of  claim 19 , wherein video fingerprints of the first hard-matching pair of video segments match, video fingerprints of the second hard-matching pair of video segments match, and video fingerprints of the classified pair of video segments in the first video content and the second video content do not match. 
     
     
         27 . The device of  claim 26 , wherein a video fingerprint of a video segment comprises an alphanumeric value, and a matching pair of video fingerprints each comprise the same alphanumeric value. 
     
     
         28 . A method comprising:
 determining a soft-matching pair of video segments of first video content and second video content, wherein the soft matching pair of video segments is located between two hard-matching pairs of video segments of the first video content and second video content; and   determining, based on the determining the soft-matching pair of video segments, a model configured to determine that a pair of video segments comprises common video content.   
     
     
         29 . The method of  claim 28 , wherein the first video content and second video content comprise different episodes of a same video program. 
     
     
         30 . The method of  claim 28 , wherein a characteristic of each video segment of the soft-matching pair of video segments does not match, wherein the characteristic comprises audio elements, an audio fingerprint, closed captioning data, subtitle data, on-screen text, or a detected visual feature. 
     
     
         31 . The method of  claim 28 , wherein common video content comprises at least one of an introduction portion, a closing portion, or an advertisement. 
     
     
         32 . A non-transitory computer-readable medium storing instructions that, when executed, cause:
 determining a soft-matching pair of video segments of first video content and second video content, wherein the soft matching pair of video segments is located between two hard-matching pairs of video segments of the first video content and second video content; and   determining, based on the determining the soft-matching pair of video segments, a model configured to determine that a pair of video segments comprises common video content.   
     
     
         33 . The non-transitory computer-readable medium of  claim 32 , wherein the first video content and second video content comprise different episodes of a same video program. 
     
     
         34 . The non-transitory computer-readable medium of  claim 32 , wherein a characteristic of each video segment of the soft-matching pair of video segments does not match, wherein the characteristic comprises audio elements, an audio fingerprint, closed captioning data, subtitle data, on-screen text, or a detected visual feature. 
     
     
         35 . The non-transitory computer-readable medium of  claim 32 , wherein common video content comprises at least one of an introduction portion, a closing portion, or an advertisement. 
     
     
         36 . A device comprising:
 one or more processors; and   memory storing instructions that, when executed by the one or more processors, cause the device to:   determine a soft-matching pair of video segments of first video content and second video content, wherein the soft matching pair of video segments is located between two hard-matching pairs of video segments of the first video content and second video content; and   determine, based on the determining the soft-matching pair of video segments, a model configured to determine that a pair of video segments comprises common video content.   
     
     
         37 . The device of  claim 36 , wherein the first video content and second video content comprise different episodes of a same video program. 
     
     
         38 . The device of  claim 36 , wherein a characteristic of each video segment of the soft-matching pair of video segments does not match, wherein the characteristic comprises audio elements, an audio fingerprint, closed captioning data, subtitle data, on-screen text, or a detected visual feature. 
     
     
         39 . The device of  claim 36 , wherein common video content comprises at least one of an introduction portion, a closing portion, or an advertisement.